On Fri, Jul 23, 2010 at 12:54 PM, Obrst, Leo J. <
lobrst@xxxxxxxxx> wrote:
> Cameron,
>
> I think our general idea was yes, OOR would be both ‘open source’ and ‘open
> system’, based on this definition. Modularity is a key requirement, and I
> think Mike Dean had made that clear from OOR’s inception.
>
>
> Thanks,
>
> Leo
> _____________________________________________
>
> Dr. Leo Obrst The MITRE Corporation, Information Semantics
>
>
lobrst@xxxxxxxxx Information Discovery & Understanding, Command &
> Control Center
> Voice: 703-983-6770 7515 Colshire Drive, M/S H305
> Fax: 703-983-1379 McLean, VA 22102-7508, USA
> From: Cameron Ross [mailto:
cross@xxxxxxxxxxxx]
> Sent: Friday, July 23, 2010 3:40 PM
> To: Peter Yim
> Cc: Adam Pease; Schoening, James R C-E LCMC CIO/G6; John Bateman; Michael
> Gruninger; John F. Sowa; Patrick Cassidy; Todd J Schneider; Natasha Noy;
> Mark Musen; Obrst, Leo J.; Mike Dean
> Subject: Re: CLIF translation of SUMO
>
>
>
> Some questions inline...
>
>
>
> On Fri, Jul 23, 2010 at 2:59 PM, Peter Yim <
peter.yim@xxxxxxxx> wrote:
>
> Thank you for the responses, Cameron. Duly noted, and I do wish you
> get funded so your good work can be open-sourced.
>
>>> [ppy] we were
>
>>> actually discussing that at the OOR meeting today (missed you there).
>>> Ref.
>>>
>>>
http://ontolog.cim3.net/cgi-bin/wiki.pl?OOR/ConferenceCall_2010_07_23#nid2FTP
>>
>
>> [CR] I think that focusing a discussion on the requirements for
>> supporting
>
>> CL within the OOR is a really good idea. I'll certainly plan to attend...
>> to
>> clarify, is this tentatively scheduled for August 19th? Not sure if its in
>> scope, but a discussion on reasoner support for CL would be really useful
>> as
>> well.
>
> [ppy] I agree "a discussion on reasoner support for CL would be
> really useful" although (especially in the OOR context) we will have
> to restrict this discussion an open source tool, as the purpose is to
> get the community to develop the *Open* Ontology Repository technology
> ... and unlike what we do at the Ontolog forum, where we can
> *occasionally* waive this IPR restriction so that we can provide
> exposure to the community as to what is the state-of-the-art ... as in
> the case of
>
http://ontolog.cim3.net/cgi-bin/wiki.pl?ConferenceCall_2010_08_05#nid2FB2
>
>
>
> I think two points are confounded here. To clarify...
>
>
>
> 1) Is reasoning in scope for the OOR?
>
>
>
> 2) Is the OOR intended to be an "open system"? By "open system" I'm
> referring to this definition:
http://www.acq.osd.mil/osjtf/whatisos.html.
> By this definition, "open source" and "open system" are distinctly
> different. The former relates to how software is licensed, the later
> relates to how software is designed and built. The source code of an "open
> system" may be open sourced, but it doesn't have to be. Likewise, lots of
> open source software is not designed and built according to this definition
> of an "open system". Open systems promote the best practice of software
> modularity (an entirely different thing from ontology modularity).
>
>
>
> Is there a consensus to design and build the OOR as an "open system"? That
> is, as a system that employs modular design, uses widely supported and
> consensus based standards for its key interfaces, and has been subjected to
> successful validation and verification tests to ensure the openness of its
> key interfaces. If so, then I believe that one of the early deliverables
> from the project would be a software architecture to support modularity
> which includes the specification of interfaces to support communication
> between modules.
>
>
> I believe that this is the way to go as it promotes a decentralized
> authority for implementing and evolving the system components etc.
>
>
> 3) Is the OOR initiative bound to architectural constraints imposed by the
> BioPortal code base?
>
>
>
> Regards. =ppy
>
> p.s. if you don't mind, we should put this out onto the [oor-forum] list.
> =ppy
> --